关于 UiPath GenAI 活动包

The UiPath GenAI Activities connector and activities allow you to work directly with UiPath-managed large language models (LLMs) from a variety of third parties. You can benefit from UiPath AI Services such as the AI Trust Layer, which gives you access to the most popular LLMs without the need to manage subscriptions, and quality-optimized, prompt-based activities with easy-to-use inputs and outputs directly in Studio Web, Studio Desktop and Apps build experiences.

Ranging from a generic completion activity called Content Generation, that allow you to define the prompt and model on your own, to predefined, use-case-specific activities like Summarize or Translate, the UiPath GenAI Activities package offers a broad range of generative AI functionalities that can be inserted directly into automations.

上下文基础

Context grounding is a component of the UiPath AI Trust Layer which allows you to bring in your data to generate more accurate, reliable GenAI predictions. Context grounding is designed to make your business data LLM-ready without the need for any additional subscription to embedding models, vector databases, or large language models (LLMs).

You can create representative indices and embeddings of business data that UiPath GenAI features can reference for contextual evidence at runtime. To learn more about its key features and use cases, see About Context grounding.

工作方式

UiPath GenAI Activities interact directly with a variety of internal UiPath services used to normalize and transform requests sent to model providers. You can create the connection directly, without any authentication parameters or subscription to a third-party service. However, you must still create and maintain the connection for use in Studio and Studio Web workflows, and Apps.

注意：您可以通过 AI 信任层 Automation Ops 策略禁用 GenAI 活动。有关更多信息，请参阅《Automation Ops 用户指南》中的“AI 信任层策略设置”。

完整的活动列表请参阅活动页面。有关更多信息，请参阅 “使用 UiPath GenAI”活动。要了解连接器，请参阅 Integration Service 用户指南中的“ UiPath GenAI 活动”页面。

查看 “支持的模型”页面，了解 GenAI Activities 使用了哪些 LLM。